Hi all
I am not a statistician but representative can be achieved with a very small (relatively speaking) sample. In the US, election polling is carried out with fewer than 2000 (like 1600) people and is a remarkably accurate predictor of outcome. This is because the strategy used for selecting the small number of informants is incredibly sophisticated.
A statistician could be more enlightening, I am sure. Point is, it is possible that the CDC sample is a more accurate representation of your population than the larger Ross sample - especially given the enormity of the population.
Cheers
Nina Berry
Australia
-----Original Message-----
From: Lactation Information and Discussion [mailto:[log in to unmask]] On Behalf Of [log in to unmask]
Sent: Thursday, October 23, 2008 10:45 PM
To: [log in to unmask]
Subject: Statistics - again
Nikki says,
<<The CDC is a survey of a representative sample of women across the country,
and is a true collection of statistics. Folks are called at home and asked
questions. The numbers for breastfeeding are not quite a straight line,
there is a very slight upward slope over the past 5 years. However,
exclusive bf rates have dropped over the past few years>>
However, gathering statistics from 424 women out of approximately 8 million
that have given birth over a two year period of time -- is that a true
representative sampling, even if they have queried all groups? I don't know how
you can get a P value out of that -- but then statistics isn't my forte -- I'll
be first to admit that. It's a sampling of 0.000053%. Ross' is a sampling
of 0.0625%.
And, if it stands to reason that Ross would benefit by having inflated
statistics, why is it that their numbers (be they from a marketing perspective or
whatever) show breastfeeding rates DECLINING where as the CDC shows them
increasing. There is a huge difference between 64% and 77% initiation rate.
While it is very satisfying to think that we have reached the Healthy People
2010 goal of 75% initiating bf, I'm not convinced we really have, because I am
having a hard time reconciling the CDC data.
Sam, you probably have seen this, but the CDC report, including the number
of mothers interviewed over the last years is at
_www.cdc.gov/nchs/data/databriefs/db05.htm_ (http://www.cdc.gov/nchs/data/databriefs/db05.htm) . You can
click on collection methods to see how many women were included.
Jan Barger, RN, MA, IBCLC, FILCA
**************Play online games for FREE at Games.com! All of your favorites,
no registration required and great graphics – check it out!
(http://pr.atwola.com/promoclk/100000075x1211202682x1200689022/aol?redir=
http://www.games.com?ncid=emlcntusgame00000001)
***********************************************
Archives: http://community.lsoft.com/archives/LACTNET.html
To reach list owners: [log in to unmask]
Mail all list management commands to: [log in to unmask]
COMMANDS:
1. To temporarily stop your subscription write in the body of an email: set lactnet nomail
2. To start it again: set lactnet mail
3. To unsubscribe: unsubscribe lactnet
4. To get a comprehensive list of rules and directions: get lactnet welcome
***********************************************
Archives: http://community.lsoft.com/archives/LACTNET.html
To reach list owners: [log in to unmask]
Mail all list management commands to: [log in to unmask]
COMMANDS:
1. To temporarily stop your subscription write in the body of an email: set lactnet nomail
2. To start it again: set lactnet mail
3. To unsubscribe: unsubscribe lactnet
4. To get a comprehensive list of rules and directions: get lactnet welcome
|